pcer_chapt4

8/8/2019 PCER_Chapt4

1/38Post-Conflict Economic Recovery: Enabling Local Ingenuity106

Billboard advertising an international money transfer company,Mogadishu, Somalia, October 2004(Sven Torfinn/Panos Pictures)

Macroeconomic Policy

Considerations inPost-Conflict Recovery


2/38

4.1 IntroductionOne of the most daunting challenges facing countries in post-conflict recovery is designing and imple-menting a macroeconomic policy regime that is appropriate to their particular circumstances.The highrisk of conflict recurrence detailed in Chapter 2 creates an additional sense of urgency. Ideally we could

visualize a virtuous circle whereby peace allows communities to regain their land and livelihoods, which

raises their incomes. Increasing incomes not only contribute to fuelling economic activity, they raise theopportunity cost of conflict relapse to potential combatants. But economic recovery is a fragile, com-plex process, and its determinants are as much political as they are economic. Should the process bederailed for any reason, political or economic, the unravelling could also be cumulative.

Political and economic stability tend to be mutually reinforcing, and yet in the immediate aftermathof war tensions often arise between political and economic stabilization objectives. For instance, largepost-war fiscal deficits may induce governments to defer salary payments, fuelling discontent andunrest.1 It is critical, therefore, for post-conflict economic policy design to recognize the politicaldimension and the incremental nature of the recovery process.

Lowering the probability of conflict relapse and consolidating peace have to be fundamental priorities inthe making of post-conflict macroeconomic policy. As discussed in Chapter 2, conflict is typically pre-ceded by a combination of horizontal inequalities, a skewed distribution of the economic surplus andresource rents, and poor overall socioeconomic performance. In other words, almost by definition, con-flict suggests that some groups have not benefited from or con-tributed to economic activity as much as they could have. Further,conflict also implies that some groups have been seriously harmedor have harmed others, creating intense mutual resentment.

Recognizing this reality, economic policy must aim not just torebuild what was destroyed by war or recreate the status quo frombefore the conflict; it must seek to break with the past.This meanspursuing policies that are conducive to more inclusive growth andenhanced human development. Stronger growth is associated withrising incomes and expanding opportunities for human develop-ment. In the post-conflict setting, economic policy must con-sciously and actively cultivate broad-based growth anddevelopment as part of the peace consolidation agenda.

Another implication of the structural, institutional and social capital damage brought by conflictis that repair will take time, and recovery must be an incremental process. Policy makers mustrank multiple high-order priorities and build on the early dividends and the promise of peace toconvince domestic and external actors to engage and invest. Such an approach will increase thelikelihood of economic recovery as a self-fulfilling prophecy. The special challenge for post-con-flict macroeconomic policy then is to kick-start the engine of economic activity, stoke it contin-uously without choking it off, and create an enabling environment for long-term change based ona cumulative strengthening of the indigenous drivers of economic recovery.

For post-conflict countries, this is a tremendous task. Conflict typically leaves countries with severemacroeconomic problems. A shrunken economic base, moderate to high inflation, chronic fiscal deficits,

Macroeconomic Policy Considerations in Post-conflict Recovery 107

Lowering the probability

of conflict relapse and

consolidating peace have tobe fundamental priorities in

the making of post-conflict

macroeconomic policy.


3/38

high levels of external and domestic government debt and low domestic government revenues are amongthe more prominent features of these economies.These factors heavily constrain the functioning of thestate and limit its ability to provide basic services, and often to cover its recurrent expenditures.The weakfiscal and current account positions signal difficulty in marshalling the resources required to jump-startthe economy. A volatile international environment, characterized by trade fluctuations and hard-to-pre-

dict aid flows, results in considerable loss of policy maneuvrability. Concomittantly, growing arrears toboth foreign and domestic creditors disqualify many post-conflict countries from immediate debt relief.

The task of breaking the so-called conflict trap is not insurmountable. There are well-known histor-ical examples of countries that enjoyed a successful recovery in the terms most stringent sense, suchas post-World War II (WWII) France and Vietnam since the 1980s.2 Even more recently, somemuch poorer war-affected countries have managed to break with the past and staged reasonable, evenimpressive recoveries. Mozambique and Cambodia are examples. Luck may have played a role, butin most cases these countries have managed creatively the instruments and resources of economicpolicy that governments have at their disposal. The lessons from successful recoveries indicate that,in addition to the usual fiscal, monetary and exchange rate policies, the post-conflict policy regimemust include employment policy and business environment policy.

These issues are the subject of this chapter. As with other dimensions of recovery described inpreceeding chapters, macroeconomic policy must be conflict-sensitive, context-appropriate andpragmatic in order to sustain recovery as well as peace. The next section describes the differentdimensions of economic recovery and highlights the diverse experiences and trajectories of 29countries that have emerged from civil conflicts since the end of the Cold War in 1989. Many ofthese countries were not performing well even before conflict; in fact, in roughly a third of themgrowth was negative before conflict. Also, during conflict, GDP per capita fell by more than 25percent in over half of the sample. Nevertheless 11 out of the 29 countries have staged fairlyrobust economic recoveries, with post-conflict GDP per capita growth averaging 4 percent orhigher.3 Perhaps more importantly, these countries grew significantly faster than before conflictstarted (Table 4.1). We recognize, however, that growth alone cannot fully define recovery. Wetherefore complement the growth story with a review of inflation and inclusiveness as key dimen-sions of economic recovery.4

The following section examines more closely the macroeconomic drivers of recovery.We look for ele-ments in the policies and the policy process that may explain the diversity of recovery experiences.

We review approaches to monetary policy and fiscal policy as well as the management of ODA. Wealso look at trade and investment facilitation, and debt management. We note broad differences withrespect to sequencing, nature and pace of reforms, as well as issues of competitiveness and policycredibility. We review the question of restoring monetary and financial oversight, and we examine themajor fiscal considerations: revenue mobilization, the tax regime and fiscal autonomy. On the expen-diture side, we examine how countries in recovery have balanced the tension between the require-ments of macroeconomic stability against sustaining social and political stability.

4.2 Economic recovery in practiceHow is economic recovery best described, given its multifaceted and multidimensional nature? Thissection attempts to capture and describe the various aspects of economic recovery as it is actually hap-pening in a group of 29 post-conflict countries for which reasonably robust data are available.

Post-Conflict Economic Recovery: Enabling Local Ingenuity108


4/38

Determining how far a country is along the path of economic recovery requires some criteria andbenchmarks to assess progress. As a first approximation, the growth of per capita output is used to geta sense of the extent to which countries are recovering. We split a set of 29 countries where conflictsended after 1989 into two groups: one group contains those countries that have experienced sustainedGDP growth per person for a significant period after conflict, the other is comprised of those that have

not.5The first group includes those countries that have met two criteria:

I post-war GDP per capita growth averages 2 percent per year or higher;6

I average post-conflict growth rate is higher than pre-conflict, hopefully indicating a breakwith the past or a transition to a new higher growth path.

The results indicate that just about two-thirds (18) of the countries for which data are available havefared reasonably well in the first few years after conflict. As introduced in Chapter 3, we refer to thisgroup as SGR (Strong Growth Recovery) countries,with perhaps half of the members of this group hav-ing clearly moved to a much stronger and sustained growth path.The other one-third (11) of countriesin our set have experienced significantly weaker recovery.We call them WGR (Weak Growth Recovery)countries.Tables 4.1 and 4.2 list the countries in each group.7


Country, start andend dates considered

Average a nnual

pre-conflict1

GDP per capita growth

Spanof

conflict

(years)

Average annual

in-conflict

Numberof

years post-

conflict4

Average annual

post- conflict

Post-minus

pre-conflict growth5

(percentage points)GDP per capitagrowth GDP per capita growth

Table 4.1 Strong Growth Recovery (SGR) countries

1 Starting from 1970 unless otherwise mentioned2 Data startin 19913 Data start in 19864 Up to20075 A positive difference indicates an improved performance over GDP per capita pre-conflict growth rates, or a potential break with the past

Source: Authors calculations based on World Bank, 2007a; IMF, 2007a; and Heston, Summers and Aten, 2006.


5/38

Figure 4.1 shows the averageGDP per capita growth tra-

jectories of both groups pre-and post-conflict. It showsthat the average per capitagrowth for SGR countries wassignificantly lower than that ofthe WGR group during con-flict (7 vs 3.1 percent onaverage). At the same time,average per capita growthrebounds more sharply andremains higher for SGRthan for WGR countries.Admittedly, such an observa-tion must be viewed with cau-tion. It could be the case thatthe post-war GDP per capitaof SGR countries grows muchfaster than that of the WGRcountries simplybecauseof thesharper drop during,and espe-cially, as Figure 4.1 shows,right beforethe end of conflict.9


Country, start andend dates considered

Average a nnual

pre-conflict1Spanof

conflict

(years)

Average annual

in-conflict

Number of

years post-

conflict2

Average annual

post- conflict

Post-minus pre-

conflict growth3

(percentage points)

Burundi 19912002 1.8% 12 -2.9% 5 0.4% -1.3

Congo, R ep. of 19931999 2.1% 7 -1.4% 8 1.8% -0.4

Cte d'Ivoire 20022003 -0.6% 2 -3.2% 4 -0.1% 0.5E l S alvador 19791991 2.0% 13 -2.5% 16 1.9% -0.2

Eritrea4

19741991 1.1% 18 -0.9% 16 1.2% NA

Guatemala 19651995 NA 30 0.9% 12 0.6% NA

Guinea-Bissau 19981999 0.4% 2 -13.2% 8 -3.2% -3.5

Haiti 19911995 -0.3% 5 - 4.9% 12 -0.1% 0.1

Namibia 19731989 -2.8% 17 -1.0% 18 1.6% 4.4

Nicaragua 19781990 1.7% 13 - 6.8% 17 0.8% -0.9

Papua New Guinea 19891996 0.7% 8 1.7% 11 -1.2% -1.9

Average 0.6% 11.5 -3.1% 12 0.3% -0.4

GDP per capita growth GDP per capita growth GDP per capita growth

Table 4.2 Weak Growth Recovery (WGR) countries

1 Starting from 19702 Up to 20073 A positive difference indicates an improved performance over GDP per capita pre-conflict growth rates, or a potential break with the past4 Until the end of the war in 1991 Eritrea is considered as part of Ethiopia and the same figures are reported for both countries

Source: Authors calculations based on World Bank, 2007a; IMF, 2007a; and Heston, Summers and Aten, 2006.

-10%

-9%-8%

-7%

-6%

-5%

-4%

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%10%

E-5 E-4 E-3 E-2 E-1 End E+1 E+2 E+3 E+4 E+5 E+6 E+7 E+8 E+9

SG R co unt ri es, ave ra ge WG R co unt ri es, ave ra ge

SGRcountries, weightedaverage WGR countries, weightedaverage

Figure 4.1 GDP per capita growth: SGR versus WGR countries8

Source: World Bank, 2007a; IMF, 2007a; Heston, Summers and Aten, 2006.

Note: The solid lines are average per capita GDP growth for all countries with data. The dotted lines are weighted average

per capita GDP growth for all countries with data (weight is share of countrys population in total groups population). See

Data Appendices for details.


6/38

Indeed, such early but transient rubber band effects have been observed in several countries includingBosnia and Herzegovina,Georgia and Lebanon,as well as Rwanda, Sierra Leone and Afghanistan,whichgrew at double-digit rates on average for the first three years after conflict. 10 However, as Figure 4.1 alsoshows, it is remarkable that when weighted by the share of population in the sample, the drop in growthrate is quite similar for both groups, but their post-conflict trajectories still diverge substantially.

The true nature of an early and sharp post-conflict growth surge can be difficult to discern if it is notsustained.11 It is noteworthy that growth remains significantly higher over a longer period for SGRcountries than for WGR countries, both in terms of simple and weighted averages.12 It could be con-

jectured that an early surge in growth brings benefits and dividends that help kick-start the recoveryprocess and put the country on a higher growth path.

The average post-war growth performance in the SGR group is 5.7 percent annually, in sharp con-trast to a near stalling (0.3 percent) on average in the WGR group over very similar average recoveryperiods of about 1112 years.13 It is worth pointing out that the average pre-conflict growth perform-ance of WGR countries was actually better than that of the SGR countries: 0.6 percent vs 0.8 per-cent (0 percent without the strongly negative growth performance of Bosnia and Herzegovina). Thedifference in their medium-term trajectories is all the more striking. There does seem to be a signifi-cant difference between the growth experiences of the members of the two groups.14

Within the SGR group, Cambodia, Mozambique, Rwanda and Uganda had impressive sustainedGDP per capita growth as shown in Figure 4.2.


7700

8800

9900

110000

111100

112200

113300

114400

115500

116600

117700

118800

119900

220000

221100

222200

EE--2200 EE--1188 EE--1166 EE--1144 EE--1122 EE--1100 EE--88 EE--66 EE--44 EE--22 EEnndd EE++22 EE++44 EE++66 EE++88 EE++1100 EE++1122 EE++1144

Cambodia, 1991, SGRMozambique, 1992, SGRRwanda, 1994, SGR

Uganda, 1991, SGREl Salvador, 1991, WGR

Guatemala, 1995, WGR

Nicaragua, 1990, WGR

Rea

lGDPpercap

ita

in2006

$Purc

has

ing

Power

Pari

ty(Base

100a

ten

do

fcon

flict)

Figure 4.2 GDP per capita in selected SGR and WGR countries (year conflict ended, group)

Source: IMF, 2007a. (estimates for 2007); Heston, Summers and Aten, 2006.

Note: Arrows point to starting year of conflict, unless where conflict is ongoing over the entire period covered (Cambodia, Guatemala), in which case the arrow points to the first year of the series.


7/38

Cambodia and Mozambique display particularly impressive growth performances, with a doublingof their real GDP per capita within 15 years after war, at an average (annualized) growth rate of 5percent and 5.5 percent, respectively. Ugandas real GDP per capita increased by 50 percent between1991 and 2007 and Rwandas by nearly 70 percent between 1994 and 2007 (at average growth ratesof 2.9 percent and 4 percent, respectively). Even with such achievements, it took Mozambique and

Uganda over 20 years (just about a generation) to reach (and, in the case of Mozambique, largelysurpass) the highest level of GDP per capita they had achieved in the five years leading up to war.As of 2007, Cambodia and Rwanda still had not quite reached that level. El Salvador, with a decent1.9 percent average annual growth rate since its 1991 peace accord (which puts it at the doorstep ofthe SGR group), has also surpassed its pre-conflict level of GDP per capita. Other countries, suchas Guatemala and Nicaragua, seem not to have been able to find a path to sustained high growth.

An important manifestation of successful recovery is obviously the absence of conflict recurrence. Onthat count, stronger growth does seem to reduce the risk of conflict relapse, since only about one infive SGR countries have experienced major conflict recurrences, while the recurrence rate for WGRcountries is about 40 percent, twice as high.15

4.3 Inflation and recoveryWhile the growth performace of countries affected by conflict varies, high inflation is typically its corol-lary. Sometimes hyperinflation occurs, as in Nicaragua where inflation reached 13,000 percent in 1987(See Data Appendices). Figure 4.3 shows that inflation in the five years before the end of war can reach

very high levels.16The averageexceeds 1,300 percent for

WGR countries three yearsbefore peace, but this is driv-en entirely by Nicaragua.Excluding this country, infla-tion is consistently and sub-stantially lower for WGRcountries in the five yearsleading to the end of conflict.

Still excluding Nicaragua, theaverage rate of inflation in

WGR countries remains lowerthan in SGR countries for sev-eral years (five or six) after theend of war, when the situationreverses with both rates fallingto single digits (Figure 4.4).Interestingly, Figure 4.4 alsoshows that the stronger earliergrowth performance of SGRcountries was achieved in spite

of high rates of inflation (70 percent one year after the end of war, when growth per capita reaches 8.9 per-cent), while WGR countries did not seem able to take much advantage of continuing reasonable inflation.


309%

78%

102%

755%

1323%

493%

9%

153%

273%

13% 17%

14%

23%

157%

62% 85%86%

0%

200%

400%

600%

800%

1000%

1200%

1400%

E-5 E-4 E-3 E-2 E-1 End0%

50%

100%

150%

200%

250%

300%

WGR countries, average (lef t axis)

WGR countries, average without Nicaragua (right axis)

SGR countries, average (right axis)

Figure 4.3 Inflation before the end of war: SGR versus WGR countries

Source: IMF, 2007b.

Note: Inflation measured as change in consumer prices (annual %). The SGR group does not include Afghanistan, Bosnia and

Herzegovina, Croatia and Georgia; the WGR group does not include Namibia.


8/38

There is also evidence of thecomplexity of the growth-inflation dynamic at thecountry level. The Republicof Congo, the Democratic

Republic of the Congo, Chadand Burundi all brought downinflation to single digits the

year after conflict ended. TheDemocratic Republic of theCongo and Chad had reason-ably strong recoveries, but inboth cases inflation was backin the double-digit range afew years later. Inflationremained low in the Republicof Congo,but the growth per-formance has been consistentlypoor. In Mozambique andUganda, among the strongestgrowth performers, inflationaveraged 50 percent and 25percent per year, respectively,for four years after conflict,

while average per capitagrowth exceeded 5 percentand 2.5 percent per year overthe same period. Unambig-uously, however, inflationabated as growth picked up(Figure 4.5).

Average inflation fell dramat-ically in SGR countries dur-ing the first five post-conflict

years (and continued to droplater). Across all post-conflictcountries, as Figure 4.6shows, increasing growth istypically associated over time

with declining inflation.Equally important perhaps,Figures 4.5 and 4.6 also suggest that high inflation rates and high growth rates are not incompatible,even though over the medium run they seem to be inversely correlated. In sum, these different resultsclearly indicate that low inflation is not a necessary or sufficient condition for high growth in the earlyrecovery period and that recovery may temporarily combine high inflation and high growth rates.


30%

23%

90%

13%

49%

3%9%

17%

8%

2.0%

9.2%

4.7%

-9.3%

5.9%

-0.7%

-8.2%

-1.8%

-4.5%

-0.3%

1.0%

-2.5%

0.0%

-10%

10%

30%

50%

70%

90%


Inflation

-15%

-10%

-5%

0%

5%

10%

15%

GDPpercapitagrowth

SGR countries, average inflation (left axis) SGR countries, average growth (right axis)

WGR countries, average growth (right axis)WGR countries, average inflation (left axis)

70%

Figure 4.4 Inflation and GDP per capita growth: SGR versus WGR countries

Sources: IMF, 2007b; World Bank, 2007a.

59%

44%

39%

7%

63%

9%

5%

0%

7%

1%

4%

11%

-0.3%

10%

-10%

0%

6%

-2%

1%2.0%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

E-4 E-3 E-2 E-1 End E+1 E+2 E+3 E+4 E+5 E+6 E+7 E+8 E+9

Ination

-15%

-10%

-5%

0%

5%

10%

15%

GDPpercapitagrowth

Uganda, growth (right axis)Mozambique, growth (right axis)Mozambique, ination (left axis)

Uganda, ination (left axis)

Figure 4.5 Inflation and GDP per capita and in Mozambique and Uganda

Source: IMF, 2007b.


9/38

4.4 Growthinclusivenessand humandevelopmentduring recoveryRecovery is likely to be moresustainable when it benefitsall, rather than only a smallproportion of the population.It is now generally recognizedthat economic growth doesnot automatically benefit thepoor. Where, for instance, themajor driver of post-conflicteconomic growth is a capital-intensive resource extractiveindustry, the benefits mayreach only a small fraction ofthe population. In Angola, forexample, as in most mineral-rich countries, investment byforeign investors in the oilsector has led to high eco-nomic growth, but thatgrowth has not yet generatedsignificant backward or for-

ward linkages to the broadereconomy nor generated sub-stantial employment.17

On the other hand, wheneconomic growth stems fromactivities that engage a largesegment of the populationand where the gains can bemarshalled to broaden thetax base and provide revenuefor services, a more inclusiveoutcome can emerge. Suchan outcome can contribute toreducing the risk of conflictrelapse because rising

incomes tend to make combat less attractivein other words, as previously noted, they raise theopportunity cost of engaging in conflict. There is some evidence that the stronger growth enjoyed bysome countries has been associated with significant inclusiveness. A growth incidence curve forMozambique, comparing 1996 (when the first major household survey took place) and 2003 shows


40%

84%

70%

46%

40%

12%

23%

7%

9%

25%

4%-3%

-1%

-2%

0%

-1%

7%

5%

7%

-2%

5%

-3%

3%2%

3%

1%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

Ination

-4%

-2%

0%

2%

4%

6%

8%

10%

12%

GDPpercapitagrowth

Average annual ination (left axis)

Average annual growth (right axis)

Figure 4.6 Inflation and GDP per capita growth in conflict-affected countries

Source: IMF 2007a.

Note: Data for Angola, Cambodia, Chad, El Salvador, Ethiopia, Guatemala, Liberia, Mozambique, Nicaragua, Rwanda and Uganda,

except inflation average excludes Nicaraguas 19861990 hyperinflation years.

Percentage of population (ordered by consumption per capita)

Averageannualgrowth

6

5

4

3

2

1

0 20 40 60 80 100

inconsumptionpercapita,percentage

Figure 4.7 Growth incidence curve in Mozambique 19962002

Source: James et al., 2005.


10/38

that growth has been both rapid and broadly inclusive,18 as the consumption of all groups increasedbetween 2 percent and 5.5 percent during the period 1996 to 2002 (Figure 4.7). Though consump-tion of the bottom 23 percent of the population grew at just over 2 percent annually, in contrast tothe 4.55 percent growth rates enjoyed by the richest deciles, this contributed to a significant declinein the prevalence of poverty, from just under 70 percent to about 54 percent.19

The main reason growthas measured by growth in consumptionhas been inclusive is its sourcein the agriculture sector, which employs the vast majority of the working population. In particular,investment by large private wholesalers in recreating grain markets has benefited food-deficit andpoorer households.20The creation of livelihoods for small producers, especially in the agriculture andrural non-farm sectors, was thus an important by-product of macroeconomic reform. Yet Figure 4.7also shows that the consumption gap between the rich and the poor widened during that time, high-lighting the fact that inclusive growth does not necessarily mean pro-poor growth (i.e. growth whichbenefits the poor disproportionately).

Ugandas mixed record on inclusive growth is also instructive. When agriculture boomed during the1990s growth was inclusive. From 2000 to 2003, agricultural growth lagged, due in part to low cof-fee prices, and the majority of the rural poor saw their incomes drop. Overall, poverty fell sharplyduring the 1990s from about 55 percent to 34 percent, then rose somewhat to 38 percent in 2003.21

Evidently, strong agricultural growth made growth more pro-poor and inclusive during the 1990s. These experiences high-light the importance of focusing on those sectors that employthe vast majority of the population if recovery is to be inclusive.

We also compare inclusiveness between our two groups usingthe World Banks Country Policy and Institutional Assessment(CPIA) ratings22 (Figure 4.8 and Table 4.3) for fifteen SGR andeight WGR countries for which they were available in 2007.23 Itappears that the SGR countries sampled tend to implementmore inclusive policies than the WGR countries sampled, witha 13 percent higher average score for the Policies for social inclusion/equity cluster.24 Furthermore,the SGR countries outperformed the WGR countries in allfive sub-categories of the cluster: SGRcountries score 18.3 percent higher for building human resources (education and health services),14.4 percent higher for equity of public resource or how public spending is distributed, 11.4 per-cent higher for social protection and labour policies (safety nets and labour market regulation), and10 percent higher on gender equity.

Again, recognizing that the direction of causalityif anyis uncertain, and that there is some con-troversy about the CPIA ratings themselves,25 an interesting inference can be drawn. In considera-tion of the stronger average growth performance by SGR countries in 2007 (around 7 percent versus1 percent for the WGR countries), these observations suggest that, at the very least, stronger growthis not hampered by more inclusive, equitable policies.

Human development

Table 4.4 complements the growth dimension of recovery by showing, where available, the humandevelopment index (HDI)26 for SGR and WGR countries.27


Inclusive recovery requires

that we focus on those

sectors that employ the bulk

of the population.


11/38


12/38

Since GDP per capita is a componentof HDI we would expect, otherthings being equal, that HDI wouldincrease faster in the SGR groupand indeed it does. Among our set of

29 countries the HDI index deterio-rates in only two countriesPapuaNew Guinea and Namibiaboth inthe WGR group. The fastest HDIincrease took place in Rwanda, withMozambique, Uganda and Ethiopiaall logging respectable performances.Chad showed the weakest HDIrecovery of the SGR group.28

4.5 Macroeconomic driversof recovery

To understand better what policieshave worked best to support recoveryin post-conflict environments, thissection reviews in greater depth theexperiences of our group of coun-tries.29 In addition to countries thatare in our list, the particular experi-ence of Vietnam will be discussed

when appropriate. The section dis-cusses five sets of policies,which col-lectively constitute pillars of macro-economic policy in post-conflict countries. These are trade- and investment-enabling policies, mon-etary policy, fiscal policy, the management of ODA flows, and debt relief.

Investment- and trade-enabling policies

Promoting an environment that fosters investment and tradeparticularly exportsis a key driverof recovery. In this section, we describe first the performance of our cohort of countries and thenreview the corresponding policy approaches that may account for the major differences observed.

Figure 4.9 compares the average export growth of SGR and WGR countries. It shows that startingat lower levels on average, post-conflict export growth increased sharply for the SGR group in thefirst two to three years (topping 20 percent as early as two years after the end of conflict). Exportgrowth remains on average significantly higher and much less volatile for the SGR than for the

WGR group in subsequent years.

Foreign direct investment (FDI) has also been crucial for recovery in a number of countries. Figure 4.10shows that FDI tends to be higher in SGR countries, and particularily so in the first few years followingthe end of conflict. The trend mimics that of economic growthwith a sharp drop at the end of conflictand massive surge post-conflictand the result is a large differential in FDI inflows between the two


11998855 11999900 11999955 22000000 22000055

Afghanistan ... ... ... ... ... ...

Angola 162 ... ... ... ... 0.45

Azerbaijan 98 ... ... ... ... 0.75

Bosnia 66 0.80

Cambodia 131 ... ... 0.54 0.55 0.60

Chad 170 0.34 0.36 0.38 0.40 0.39

Congo, Dem. Rep. ofthe 168 0.43 0.42 0.39 0.38 0.41

Croatia 47 0.81 0.81 ... 0.85

Ethiopia 169 0.31 0.33 0.35 0.38 0.41

Georgia 96 ... ... ... ... 0.75

Lebanon 88 ... 0.69 0.73 0.75 0.77

Liberia ... ... ... ... ...

Mozambique 172 0.29 0.32 0.34 0.38 0.38

Rwanda 161 0.40 0.34 0.33 0.42 0.45

SierraLeone 177 ... ... ... ... 0.34

SolomonIslands 129 ... ... ... ... 0.60

Tajikistan 122 0.71 0.70 0.64 0.64 0.67

AAvveerraagge 113322 00..4411 00..4499 00..4499 00..4488 00..5577

Burundi 167 0.35 0.37 0.35 0.37 0.41

Congo, Rep.of 139 0.57 0.56 0.55 0.52 0.55

Cte d'Ivoire 166 0.45 0.45 0.44 0.43 0.43

El Salvador 103 0.61 0.65 0.69 0.72 0.74

Eritrea 157 ... ... 0.44 0.46 0.48

Guatemala 118 0.57 0.59 0.63 0.67 0.69

Guinea-Bissau 175 0.30 0.32 0.35 0.37 0.37

Haiti 146 0.46 0.47 0.49 ... 0.53

Namibia 125 ... ... 0.70 0.66 0.65

Nicaragua 110 0.60 0.61 0.64 0.67 0.71

PapuaNewGuinea 145 0.48 0.50 0.53 0.54 0.53

AAvveerraaggee 114411 00..4499 00..5500 00..5533 00..5544 00..5555

SGR countries

Countries

WGR countries

HDI HDI HDI HDI HDIRRaannkk22000055

Uganda 154 0.42 0.43 0.43 0.48 0.51

and Herzegovina

Table 4.4 Human development index: SGR and WGR countries

Source: UNDP, 2007a.


13/38

groups in the early (three tofour) years after conflict whenthe economy is in great need ofresources and peace dividends.Over the medium term, both

groups seem to converge tosimilar levels, before beginningto diverge again.

The experiences and approach-es of recovering countries todoing business have dependedon their individual characteris-tics and circumstances. Somenatural resource-rich coun-tries, such as Mozambique,have been highly attractive toFDI.30 The evidence showsthat FDI favours resource-rich countries and sectors.31

However, beyond mereresource endowment, there isalso evidence that policiesmatter. Mozambiques relativepolitical and macroeconomicstability, a relatively liberaltrade regime, cheap labour andsound exchange rate manage-ment have played an impor-tant part in its performance.Cambodia, where oil was dis-covered only recently,32 had abig surge in apparel exports inthe five years after the war.33

Similarly, Vietnams impres-sive performance in the 1990shad little to do with mineralsand metals.

In the immediate aftermath of conflict, companies are reluctant to invest because of continuing secu-rity, political and business risks. Protracted physical insecurity, widespread corruption, cumbersome orinefficient legal and regulatory systems, destroyed infrastructure and a poor stock of human capital allcontribute to a poor investment climate. Low or uncertain financial returns deter the engagement ofprivate enterprise. As noted above, the main exceptions are in natural resources and, as noted inChapter 3, sectors such as telecommunications. For most countries, however, in the immediate after-math of conflict the bulk of financial inflow typically comes from foreign aid; significant private


-10%

-5%

0%

5%

10%

15%

20%

25%


SGR countries, average

WGR countries, average

Figure 4.9 Export growth: SGR versus WGR countries

Source: World Bank, 2007a.

Note: Value of exports of goods and services (constant 2000 US$), calculated as percentage change. The SGR group does not

include Afghanistan, Angola, Liberia, Sierra Leone and Solomon Islands; the WGR group does not include Burundi, the

Republic of Congo, Haiti and Papua New Guinea.

0

1

2

3

4

5

6

7

8

9


AsapercentageofGDP



Figure 4.10 FDI: SGR versus WGR countries

Source:World Bank, 2007a.

Note: The SGR group does not include Afghanistan; the WGR group does not include Namibia.


14/38

investment comes in only several years after conflict has ended. Getting private enterprise to enterpost-conflict economies significantly earlier remains a continuing challenge that requires additionalcreativity to raise the financial incentives for companies to invest in key sectors (Box 4.1).

Two lessons may be drawn from the review of business-enabling policies. First, most successful growth

recovery countries have implemented reforms early on, yet with a gradualist and sequential approach.Bymost accounts, countries such as Mozambique and Uganda have implemented orthodox reforms,focused on macroeconomic stability and structural reforms, with significant donor support. For politicaland historical reasons, Vietnams and Cambodias reforms have had a more unorthodox flavour, espe-cially in the former case with a tightly state-led recovery. And yet,far from implementing shock therapy, all these countries haveadopted a sequential, step-by-step approach, with early, often eas-ier, reforms facilitating and laying the ground for those that follow.

Policy makers in Mozambique initially enacted foundationalreforms, which were relatively easy to implement and which pro-

vided the basis on which other reforms could be built. 34These earlyreforms focused on overcoming fiscal crises, curbing hyperinfla-tion35 and achieving basic macroeconomic stability. Once the foun-dation was laid, more ambitious reforms in the banking system,investment regime and public finance became easier to implement.

This sequencing of reforms not only strengthened the hand ofreformers in the government but also allowed the building of thepolitical will and institutional capacity that would be needed laterto design and implement more difficult reforms. Vietnam also presents an interesting case study for astep-by-step economic reform process tailored to indigenous needs and realities. For instance, its tradeliberalization followed a two-stage strategy,with a gradual reduction in tariffs and non-tariff barriers thatgave local firms time to adjust and strengthen before being fully exposed to international competition.In Uganda, economic reforms broadly followed three phases: stabilization and recovery measures, struc-tural adjustment and liberalization, and structural and institutional reforms. A large economic recoveryprogramme focusing on stabilization and devaluation measures, coupled with essential investments ininfrastructure rehabilitation, was followed by price and structural reforms, including the legalization ofthe parallel exchange market and the abolition of the state monopoly on coffee marketing. Finally, ambi-tious reforms were introduced in the 1990s to strengthen the public financial management system,restructure the banking sector, and improve revenue collection. As a result, and helped with significantdonor support, Uganda was able to regain investors confidence.36

In addition to their economic benefits, the gradualist approach used by these countries further allowedthem to build the self-confidence and general goodwill that could then be deployed for introducing morecomplex reforms. To regain the confidence of the population and rebuild trust after years of uncertaintyand violence requires the provision of jobs and basic social services for individuals and households. Forcommunities and local government it means revenue sharing, institutional strengthening and relative fis-cal autonomy; for business it means political stability, clearly articulated economic policies and a commit-ment to rebuilding investment-enabling institutions, especially financial networks and key public services.

The lesson is that the gradualist approach allows recovering economies to build on early dividends in aself-reinforcing manner.This means that the early years are critical in determining the recovery path.


Successful growth recovery

countries have in general

adopted a sequential, step-by-

step approach, with early,

easier, reforms facilitating and

laying the ground for those

that follow.


15/38Post-Conflict Economic Recovery: Enabling Local Ingenuity120

BOX 4.1 ENABLING THE PRIVATE SECTOR IN POST-CONFLICT ECONOMIES

Considerations to foster private sector involvement in post-conflict settings include:

I Initiate reforms as early as possible. There is a relatively narrow window of opportunity to establish a new status quo that is

conflict-sensitive, more inclusive and which promotes economic transformation. It is critical to embark on such reforms in the period immedi-

ately following the end of a conflict, before the possible entrenchment of opposing political interests.

I Prioritize governance and security reforms. Ensuring physical security, property rights protection and transparency in the

justice system are central to improving the business climate and attracting private investment anywhere, but particularly so in the high-risk

environments of post-conflict economies. Laying the basis for these immediately sends a powerful signal that may raise the confidence of

domestic and foreign investors alike in the future of the economy.

I Empower local entrepreneurs. Supporting local business development is essential to economic recovery as the revitalization

of domestic enterprise signals the return of normalcy, in addition to restoring local markets and providing employment. Key steps for engaging

the local business community include improving its access to finance, through such mechanisms as microcredit and, even more importantly,

encouraging its continuing participation in policy dialogue with the government and policy makers.

I Promote foreign investment. Perceptions of risk delay entry of private enterprises into post-conflict environments. For exam-

ple, extractives industries may be quick to pursue exploration contracts in resource-rich countries, but may delay long-term investments in pro-

duction until they are better assured of a countrys stability. Incentives such as tax allowances and political risk insurance (PRI) may encourage

foreign investors to enter the post-conflict market earlier. The World Banks Multilateral Investment Guarantee Agency (MIGA) could be an

important mechanism to help reduce the hazards of investing in unstable markets. A donor-funded investment guarantee facility that extends

PRI access to domestic investors has been proposed in the report of the Commission for Africa. Extending political risk coverage to domestic and

regional investors may encourage them to invest earlier than they were previously considering;

I Promote publicprivate partnerships (PPPs). PPPs allow the private sector to deliver services previously (or traditionally)

provided by the public sector. In the post-conflict setting, the model could be used creatively, particularly in restoring essential infrastructure

and rehabilitating schools and health facilities. In addition to the direct recovery dimension, PPPs could help grow domestic private enterpriseif special efforts are made to link the main private leader to domestic suppliers and subcontractors.

I Be realistic about investment opportunities. Individual company calculations for investing vary according to their size, sec-

tor and corporate strategies. While participation of larger, transnational corporations companies may represent a strong vote of confidence for

local stability, investment from these entities may not be realistic in the short term. Policy makers may want to focus on smaller, local or region-

al actors who have greater familiarity with local conditions and who may have a greater stake in local economic recovery. The reasons for such

additional interest from regional enterprises may well include strong cultural linkages.

I Develop strategies to enhance risk sharing. Recovering economies need a robust risk-sharing facility to help bridge the gap

between private and social returns that characterize post-conflict situations. While private returns to the investor may be low, the social returns

are usually considerable. These include the signalling that peace has returned that may attract in additional investment, and the employment

generated that may reduce conflict risk. This divergence between private and social returns suggests an important market failure correction role

for a publicly funded financial facility. The scope of such a facility could include bearing the cost of doing project design and financial structur-

ing in a fully consultative and conflict-sensitive manner. It could also bear some of the cost of borrowing as an incentive for early entry into high-

priority sectors such as energy, construction or export industries.


16/38

Second, privatization and financial sector reforms were often part of the wider pro-business reformpackage, with varying results. Many post-conflict countries have engaged in large-scale privatizationprogrammes to sell state assets to private entrepreneurs. In Mozambique, Bosnia and Rwanda, forinstance, external partners made privatization an important conditionality in the policy reform.37 InLebanon, a wide range of enterprises and public utilities were transferred to the private sector.38

Given the shortage of public funds, privatization presents a practical way to raise resources.Privatization has also been promoted as an instrument to stimulate the private sector and initiate thecrucial mobilization of domestic and foreign resources.

In practice, however, the record of post-conflict privatization has been mixed as a result of a varietyof obstacles, even in strong performing countries like Mozambique, where privatization was at theforefront of economic reforms. Privatization was often undertaken without sufficient preparation and

without having installed an adequate regulatory, financial and competitive framework. In some post-conflict economies with flourishing criminal networks, privatization has frequently been a garagesale for the personal enrichment of corrupt business interests or, as in Sierra Leone, taken over bypowerful elites, and thus failed to bring about an efficient market.39 Similarly, according to one ana-lyst, six of the 14 state-owned banks operating in Bosnia and Herzegovina were informally priva-tized during the war, including the transfer of assets to war criminals.40 In other cases, as withMozambiques largest bank, privatization was essentially a sale to foreign companies, raising concernsabout national sovereignty.

Against this situation, some governments have opted to encourage private participation in infrastruc-ture provision and public service delivery, rather than sell assets. In some situations, governments haveencouraged competition between the public sector and the private sector. Ethiopia, for instance, intro-duced private banks side by side with state banks to induce greater competition in the banking sector.In Cambodia also, hundreds of tiny private power networks established themselves throughout thecountryside during and after the civil strife of the 1990s, effective-ly filling the void left by the non-functioning national utility.41 Inthat situation, the government chose wisely to do nothing toimpede new entrants into the business. In Vietnam, an effectiveand viable approach was found to be the restructuring of publicenterprises and bringing them under market discipline.

Policy implications

There are many ways to improve the business and investmentclimate, and some approaches are more orthodox than others.In all cases, however, there was no doubt that the governmentmeant business. This was demonstrated in efforts to restorecredibility, confidence and competitiveness. Domestic resourceendowments and conditions play a role, but commitment andpolicies seem to matter more.The sequencing and context-appropriateness of these reforms seem togo a long way, casting doubt on the benefits of shock therapy in restoring an enabling environment.Gradualist approaches strengthen the feasibility of recovery, with early dividends paving the way forfurther reforms in a self-reinforcing manner. In resource-rich economies, FDI and exports could pro-

vide the basis for such dividends. In less-endowed environments the role of donors in kick-startingthe recovery process is likely to be critical. In all cases, creativity is required of the state.


Government commitment

to restoring credibility,

confidence and

competitiveness is paramount

for improving the business

and investment climate.


17/38

Re-establishing a credible and efficient monetary policy regime

Rebuilding the domestic financial and banking sector to protect depositor interests and help chan-nel resources between savers and investors has been one of the greatest challenges of post-conflictrecovery. Weak supervision and regulation of the sector is typically a by-product of under-resourcedand understaffed central banks. This may result in regulatory forbearance under the pressure of pow-

erful interests, including war criminals, leniency in the licensing of banks and general failure to curbemergent bank crises.42 It is critical, therefore, to re-establish a functioning central bankor itsequivalentas a first step towards strengthening the financial and banking sector.

Re-establishing the central bank is also pivotal for managing monetary policy and exchange rates in ways thatthat will help accelerate and sustain recovery. In environments where bartering may have been substituted forcurrency transactions, the government must enact policies to restore the credibility of the monetary systemand its authority over it. In parallel, post-conflict governments need to meet their financial obligations,including paying employees, pensioners and suppliers. A credible monetary policy must thus balance theneeds for price stability with the need for liquidity to lubricate the economy and jump-start development.

The indigenous drivers approach suggests that the starting point for rehabilitating the central bankshould be to assess the existing institutional capacity and build upon it. In countries like Rwanda wherethe central bank shut down during the war, the first challenge was to reopen it. In Afghanistan, amidstdifficult circumstances, a strong effort by the government successfully led to the central banks introduc-

tion of a local currency that facilitatedeconomic management and helpedstate-building (Box 4.2).

In Afghanistan, Bosnia and Herze-govina, and Croatia, the timely over-haul of monetary policy reduced thecontribution of price instability toaggregate uncertainties, while in

Tajikistan any beneficial effects ofpost-conflict currency replacement

were eroded by the return of mone-tary indiscipline. 43

Two types of monetary regimes havetypically been used in post-conflict set-tings, because of their strong commit-ment to price stability and lowtechnical requirements from the cen-tral bank.The first, the use of a foreigncurrency (usually dollarization), doesnot carry the requirements of a centralbank and is immune to governmentinterference. Dollarization, however,has drawbacks because it limits thecountrys ability to cope with external


BOX 4.2 MONETARY RECONSTRUCTION IN AFGHANISTAN

After the 2001 war, there was a pressing need in Afghanistan to rebuild the econ-

omy and strengthen key institutions. Years of civil war had left Afghanistan with a

shattered economy, a weak central bank, no local currency and an ineffective pay-

ments system. The economy was heavily dollarized, and there was a proliferation

of foreign currencies as well as counterfeit ones. Under these circumstances, the

government decided to introduce its own currency so it would be easier to meet its

financial obligations, enhance state power and improve economic efficiency.

Moreover, with a new currency, confusion could be eliminated. During a nine-

month period, the government hired a reputable German printer to print the new

notes and developed and implemented complex logistical plans for the safekeep-

ing and distribution of 500 tons of new notes throughout the country. Through

buying and selling dollars, the central bank created liquidity during the transition

period. The moneychangers, the hawala, were brought into the exchange process

and were very important for its success. As a result, in less than a year the govern-

ment had a reliable currency for budgetary operations and an important symbol

for state-building success.

Source: Coats, 2007, p. 237238.


18/38

adverse shocks. Dollarization may protect the country from financial instability, but it deprives nationalauthorities of control over monetary policy. It can also have a negative wealth effect, because the wealthof poor people who hold local currency declines as the currency depreciates with respect to the dollar. 44

A country choosing to use a foreign currency is essentially giving up the revenue it can collect fromseigniorage. Full dollarization in Liberia would have increased the economys vulnerability to external

shocks in addition to removing the governments revenue from seigniorage.45 Monetary managementbecomes more difficult since the central bank has very few levers to influence the supply of money. Dueto these weaknesses of dollarization, most of the recent post-conflict economies have favoured keepingtheir own currency.

A second option is a currency board arrangement under which a country pegs its exchange rate to aforeign currency and backs its money supply with foreign exchange. Such an arrangement allows thecentral bank to capture the seigniorage and retain the political benefits of a national symbol. 46 Oneplace where a currency board has been successful in reducing inflation and preventing the central bankfrom printing money has been Bosnia and Herzegovina. In this case, the fact that discretion over mon-etary policy was not assigned at the Dayton Peace Accords facilitated acceptance of the Accords bythe Bosnian, Croatian and Serbian representatives.47 Reserves in the Bosnian Central Bank more thandoubled between 1995 and 2000, the first five years of recovery, and inflation fell significantly.48

In economies experiencing hyperinflation (inflation rates of 50 percent or more on a monthly basis),partially fuelled by the monetization of fiscal deficits, tight fiscal and monetary instruments may be theonly available remedy. In the Democratic Republic of the Congo, breaking a vicious cycle of hyperin-flation and falling currency required a decisive stabilization policy involving a major tightening of thefiscal stance.49 In some countries, soft pegs may help anchor the currency and give greater operationalflexibility to the central bank, but there will be considerable pressure on the central bank to manage theexchange rate regime effectively and withstand political pressures to lend to finance governmentdeficits. A successful monetary and exchange rate regime in a post-conflict country will require a cen-tral bank that is substantially autonomous and immune to political pressures to fully fulfil its role.

Good exchange rate management has also proved to be essential in post-conflict countries for main-taining trade competitiveness. An appreciation of the currency is a real concern in a poor rural econ-omy where agricultural exports are important (such as coffee exports in Ethiopia, Rwanda andUganda), and where a fall in the domestic value of these exports can have serious consequences forrural poverty and employment. In addition, a short-term signal to firms that traded goods are lessprofitable may discourage or delay critical investments in export and import-competing industries.

There is some empirical evidence of the economic benefits of preventing an inflation-adjustedexchange rate appreciation in order to maintain competitiveness.50

Figure 4.11 shows that the SGR group (with stronger export growth performance than the WGRgroup) benefited from an increase in their trade competitiveness. The loss of competitiveness couldalso be a particular concern for post-conflict countries that receive large aid inflows, which couldpotentially lead to currency appreciation as discussed later in this chapter.

Policy implications

This review of post-conflict monetary policy experiences suggests that prompt, well-executed andpublicized monetary policy overhaul is beneficial for reducing uncertainty, facilitating assistance and



19/38

establishing that the orien-tation of the post-conflictgovernment is towards thesocial good.51 It also showsthat there are several dif-

ferent ways to reach similargoals and that the choice ofan appropriate monetaryregime depends on theeconomic fundamentalsand institutions of theeconomy.

Fiscal policy

The main challenge forpost-conflict fiscal policy isto find the right balancebetween competing exi-gencies, at a time whenneeds are great and capaci-ties are very weak. Giventhe high risk of conflictrecurrence, fiscal policymust support the consoli-dation of political stability,

while not jeopardizingmacroeconomic stability.

The high inflation andindebtedness that are typi-cal of post-conflict coun-tries reduce the scope forfurther deficit financing.Post-conflict countries alsousually display very lowgovernment revenue/GDPratios,53 as a result of weak-ened capacities for tax col-lection and the widespread

war-induced informality that reduces the tax base (Figure 4.12). Furthermore, like most developingcountries, they have greater reliance on indirect taxesespecially trade taxesthan on direct taxes.

While the revenue/GDP ratio tends to follow an upward trend after conflict, domestic revenuesremain largely insufficient to meet recovery needs.

The first consideration in restoring the fiscal capacity of the post-conflict state is the re-establish-ment a proper and functioning tax administration. This is discussed in Chapter 5. The second con-


0 10 20 30 40 50

Afghanistan

Congo, Dem. Rep of the

Sierra Leone

Liberia

Low income countries

Timor-Leste

OECD countries

European Union (EU)

As a percentage of GDP

Total domesticr evenue

of which revenue from trade taxes

(except for EU, Timor-Leste andAfghanistan)

Figure 4.12 Government revenue in selected countries and groups of countries

Source: Carnahan, 2007, Figures 1 and 2, p.2.

50

60

70

80

90

100

110

120

E-8 E-7 E-6 E-5 E-4 E-3 E-2 E-1 End E+1 E+2 E+3 E+4 E+5 E+6 E+7 E+8 E+9

Index

(Base

100

ate

nd

ofconict)


WGRcountries, average

incr ease in 'price' competitiveness

Figure 4.11 Trade competitiveness index: SGR versus WGR countries52

Sources: Berg, 2006; IMF, 2004b; IMF, 2007e; IMF, 2007f; IMF, 2008a; World Bank, 2007a; World Bank,2001.

Note: This is the trade weight real exchange rate computed as domestic consumer prices divided by a trade weighted index of

whole prices of trading partners (a fall is a depreciation of the real exchange rate).


20/38

sideration is reform of the tax system. Corresponding to the overall approach of this report, reformshould be guided by pragmatism, gradualism and conflict sensitivity. Pragmatism suggests that suchreforms should start from whatever system and taxes are in place. This might, for instance, entailreducing the number of tariff rates and classifications, as in Afghanistan, where 25 tariff rates thatranged from 7 to 150 percent were replaced with a simplified six-rate structure.54 Pragmatism is also

manifest in the case of the IMF-supported introduction of a levy on coffee export in Timor-Leste,although in some other cases orthodoxy has prevailedin Guatemala for instance, though revenueenhancement was included in the peace accords, priority was given to cutting tariffs.55

Gradualism implies the momentary toleration of second-best policies that permit revenues to bemobilized quickly while the economy expands and capacities are rebuilt. Practices such as withhold-ing taxes on wages (as in Afghanistan, Bosnia and Herzegovina and Timor-Leste) and a presump-tive tax on small business (in Timor-Leste) can help lay the foundation for a more efficient personaland corporate tax system in the future. In Timor-Leste, the export tax on coffee was subsequentlyreplaced by a general income tax as the recovery progressed. In Mozambique, the Government, withsupport from the IMF and DFID, contracted a private firm (Crown Agents) for a build, operate,and transfer system that resulted in increases in revenue mobilization from customs.56

Finally, and perhaps most importantly, reforms to the tax system must be conflict-sensitive. Theirimpact on vertical and horizontal inequalities must be explicitly considered. Even as we consider theusual questions of revenue-generating potential, administrative feasibility and efficiency, we mustalso recognize the imperative of keeping the risk of conflict recurrence low.

The third aspect of restoring fiscal capacity has to do with the size of the tax base. This dependsprimarily on the size and dynamism of the taxable formal economic sector. In the aftermath of con-flict, it is critical to re-establish the control of the state over the territory and the corresponding eco-nomic activity. It is particularly important to rebuild and expand activity in the formal sector. Thisrequires not only the use of taxes as incentives, but also the use of government expenditure alloca-tion to the same end.

The level and pattern of public expenditures affect the prospects of recovery in many ways. Meetingbasic needs offers a tangible peace dividend for war-torn societies, and is central to state legitimacy andthe social contract between the state and society.57 Public expenditure can be the initial main engine ofgrowth and development and is the main source of the peace dividend,which is essential to peace con-solidation. But if public expenditure increases too rapidly to meet such urgent priorities as employmentcreation, large fiscal deficits and inflation may follow. There is some evidence that in post-conflict set-tings, each incremental episode of inflation leads to additional capital flight (or currency substitution)undermining domestic seigniorage revenues and potentially destabilizing domestic financial markets. 58

Fortunately moderately high inflation may not hamper growth, as discussed above.59 Still, a middleground needs to be found between the imperatives of macroeconomic stability and the imperatives ofmaintaining peace. For post-conflict countries, the short run is particularly important. The macroeco-nomic goal of price stability and the peace-building goal of political stability must be pursued togeth-er. The former requires efforts to minimize budget deficit while the latter may require specialexpenditures to implement peace-accord commitments or to meet urgent social needs. In such settings,macroeconomic orthodoxy is impractical.60 In Guinea-Bissau, wage arrears have contributed to pro-found unrest and undermined administrative effectiveness. In the Central African Republic,61 the



21/38

accumulation of large arrearsand the irregular payments of

wages to civil servants andmilitary personnel have led toprolonged labor strikes and

army mutinies.62 On theother hand, in some cases,high inflation can also createunrest.63

Figure 4.13 shows that SGRcountries ran larger publicsector deficits than WGRcountries. This suggests thatthere may be some room torun relatively high post-con-flict public sector deficits.

Public expenditures can be recovery enabling if they are able to effectively address horizontalinequalities, promote inclusive growth and generate employment. Some countries have been rathersuccessful at this. Cambodia combined additional Multi-Fiber Arrangement (MFA) quotas for gar-ment exports with an ILO-run factory certification programme to create a very rapid increase inmanufacturing employment, particularly for young women (Box 4.3). El Salvador managed to trans-form its rural, employment-generating social investment funds into a system of cash transfers target-ing women in the poorest villages.64

The evidence from several post-conflict experiences suggests, however, that this is seldom achieved.For instance, a study of the distributive incidence of post-war public expenditure in Guatemalafound a persistence of both vertical inequalities and horizontal inequalities between Guatemalasindigenous and non-indigenous populations.65 Further, in comparison to other countries at similarlevels of income, the budget allocations for health expenditure, as well as for secondary and tertiaryeducation, were regressive.

And yet, given the importance of horizontal impact of expenditure inequalities as risk factors of conflict,policy makers must ensure that public expenditures contribute to enhancing the well-being of the less-privileged regions and groups so as to alleviate grievances. At the very least, it must not be regressive. As

with revenue mobilization, this requires a sound analysis of the distributive impact of expenditures.66

Policy implications

Two main policy implications may be drawn from the discussion of fiscal policy. First, the distrib-utive impact of fiscal policy and its effect on conflict dynamics must be central considerations indeciding on revenue and expenditure policies. Such a concern might call for decisions regardingdeficit targets, tax reforms or expenditure programmes that differ from those that would have pre-

vailed in non-conflict settings. Second, timing and sequencing are particularly important. A key pri-ority in the immediate aftermath of conflict is to create economic and employment opportunitiesand to start rebuilding the capacity of fiscal institutions as quickly as possible. One of the most crit-


-16

-14

-12

-10

-8

-6

-4

-2

0

2


Asapercentag

eofGDP



Figure 4.13 Public sector deficit without aid financing: SGR versus WGR countries

Sources: Beresford et al., 2004; Berg, 2006; IMF, 2004b; IMF, 2007e; IMF, 2007f; IMF, 2008a; World

Bank, 2007a; World Bank, 2001.

Note: The SGR group does not include Croatia, Lebanon, Liberia and Solomon Islands, the WGR group does not include Haiti.


22/38

ical needs after war is restoringthe social contract between thestate and the citizens, even ifdealing with a transitionalauthority. In such circumstances,

the first order of business shouldbe for the state to regenerategoodwill via the provision ofpublic goods and the delivery ofsocial services. Such an approach

will also provide significantassurance to the populace thattheir tax payments (when theyare imposed later) will be put togood use. Too much focus tooquickly on taxing the populationmight be counter-productive,especially because the tax basehas been so severely eroded byconflict. Rebuilding the institu-tions for revenue mobilization

will lay the groundwork for sub-sequent increases in tax revenuesas the economy recovers.67

Official developmentassistanceBeyond the private sector andgovernments, ODA constitutes apillar of recovery for most post-conflict countries today. To gainsome insight as to whether andhow ODA can support economicrecovery, we start by drawingsome lessons from the experience of the first modern post-conflict reconstruction package, andarguably the first ever large-scale external assistance programmethe European RecoveryProgramme after World War II, known also as the Marshall Plan.

Most importantly, the Marshall Plan did not help restore the pre-war economy of Europe; rather ithelped Europe break with the past and move on to a new, higher growth trajectory (see Figure 4.14for France and Italy) supported by stronger and wider social safety nets. It also helped alleviate thebitter nationalist and ethnic rivalries that had driven centuries of devastating wars in Europe.

There are clearly important differences between contemporary countries in post-conflict recovery andItaly, France and Germany in the aftermath of World War II. The latter ranked among the most devel-oped nations in the world in 1939, and had solid institutions and high stocks of human and physical cap-


BOX 4.3 THE ILOS BETTER FACTORIES CAMBODIA PROGRAMME

Ensuring that some labour standards are observed is an important consideration both for

the health and safety of local workers and for the reputation of companies operating in

fragile post-conflict environments. International interventions may be directed to help put

in place an adequate regulatory framework. In 2001 the United States and Cambodia

entered a three-year trade agreement increasing Multi-Fiber Arrangement (MFA) quota

access to the US clothing market. The government of Cambodia asked the ILO to develop a

factory working conditions certification programme to make sure labour laws were

enforced in its garment factories. In response, the ILO initiated the Better Factories

Cambodiacertification programme.

The garment sector is critical to Cambodias economy and employs a substantial portion of

the workforce, the majority of which are young women: at the start of the programme, an

estimated 90 percent of the workers in the industry were young women, mainly from rural

areas, and 85 percent of them were under 26. The ILO hired and trained its own factory

inspectors to monitor and report on working conditions and to help companies bring theirfactories in line with Cambodian labour law and ILO standards.

The programme has since been cited by large corporate buyers as an inducement to use

Cambodian factories. As a result, exports expanded rapidly and nearly a million jobs have

been created by these factories. Employment of women in the industry grew at nearly 20

percent annually between 1993 and 2001 (albeit from a very low base) as the industry

grew from 5 percent to almost 12 percent of total employment during this period. By

some estimates one in eight Cambodian families have at least one member at work in a

garment factory.

By raising its labour standards Cambodia was able to increase its market access. This sort

of donor-supported certification and inspection programme is especially useful in war-

torn countries with weak technical capacities for oversight and enforcement.

Sources: ILO, 2005; ILO, 2006; Becker, 2005; Polaski, 2006.


23/38

ital. Most of the former arepoor rural economies domi-nated by agriculture andresource industries, perhaps

with the exception of

Bosnia and Herzegovina,Croatia and Lebanon. How-ever, when the MarshallPlan was launched in mid-1947, Europes devastatedeconomies were sufferinghigh inflation, rationing,social unrest, and even ashort but violent civil war inItaly. Two years later, infla-tion had fallen back into

single digits, budget and trade deficits had narrowed and investment and productivity growth began aremarkable 30-year run.68

What lessons can be drawn from the design and implementation of the Marshall Plan? First, theeffort and resources that went into the Plan were considerable. Financial input ranged betweenUS$20 and US$30 billion a year at todays prices over four years. 69This is equivalent to 30 percentof the total ODA provided by all major donors in 2007.70 Its ambition went beyond restoring a pre-

war order; it aimed for a qualitatively different socioeconomic order.71The volume of aid providedmuch-needed fiscal space for Europe, allowing the countries to finance the reconstruction of keyphysical infrastructure, to extend social safety nets, to ease shortages of basic necessities and to relaxprice controls and rationing. This helped governments regain some popularity and boosted confi-dence that recovery was on track.72

Second, its predictability made medium- and long-term planning feasible, and enabled consultationamong different political factions within beneficiary countries. This helped buy time and policyspace to build political consensus, including on the role of private and public enterprise.74 Franceand Italys governments were in delayed stabilization mode: divided political factions blocked fis-cal and price reforms fearing their constituents would be hurt. By reducing the share of the budgetdeficit financed locally, easing the distributive conflicts that drive up wages and prices and thusrelaxing the urgency of sharp adjustment, and by compensating potential losers, Marshall Planfunds helped break the deadlock.75

Third, limited conditionality enabled context-appropriate decisions to be made with respect to theabsorption and spending of aid. Critically, for instance, as discussed in Box 4.4, both France and par-ticularly Italy saved some of the funds they received to build up reserves, lower debt and help stabi-lize their exchange and inflation rates. Not absorbing aid reduced pressure on the exchange rate toappreciate. By 1950 Italys trade plus services deficit shrank to just US$47 million, though 1950 aidflows were US$239 million. In part because inflation fell, financial stability improved and key exportindustries flourishedItalian textiles for exampleand the political coalition assembled to secureMarshall Plan funding proved durable.76


1,000

10,000

100,000

1869

1874

1879

1884

1889

1894

1899

1904

1909

1914

1919

1924

1929

1934

1939

1944

1949

1954

1959

1964

1969

1974

1979

1984

1989

1994

1999

LogGDPpercapita(1990international

Geary-K

ham

isdollars)

Italy

France

Figure 4.14 France and Italys GDP per capita 1869-1995

Source: Maddison, 2003.


24/38


25/38

high during conflict, butlow after conflict ended.However, since 1989 therehas been a tendency for aidinflows to shrink during

conflict, to increase sharplyin the first few years afterconflict, and to taper offlater on.

There is, however, signifi-cant heterogeneity acrosscountries, with importanteconomic and policy impli-cations (Figure 4.15).

ODA to post-conflict coun-tries has tended to favour SGR countries, such as Afghanistan, Bosnia and Herzegovina,Mozambique and Uganda, as opposed to WGR countries like the Republic of Congo, Guinea-Bissau or Burundi.

By and large, aid as a share of gross national income (GNI) is very similar for the two groups beforeand during war, and then again between five years and later after war ends, with a ratio in the rangeof 1015 percent (and more stable levels for WGR countries).The main divergence occurs during thefive years immediately after the end of war, with SGR countries enjoying a much more pronouncedincrease in aid as a share of their GNI at a time when they also enjoy on average, higher growth ratesthan WGR countries. The overall difference is large: six percentage points per year on average for five

years. For African countries in particular, this difference in the level of aid raises important questions,given that, as previously noted, roughly the same proportion of countries (half of them) in each of ourgroups are African (6 of 11 in the WGR group and 9 of 18 in the SGR group).

The pattern revealed by Figure 4.15 does not necessarily imply that aid causes rapid economic recov-ery, as causality could in fact run the other way: successful and peaceful recoveries (following severe

wars) attract more donor attention. Aid may also, by design, follow more successful peace accords orUN intervention, which would then be the determining factor. However, the data suggest that a bigspike in aid immediately after conflict ends may contribute to strong growth performance, and atleast does not impede it. Many countries that have out-performed their pre-conflict growth rateshave done so in an aid-abundant environment. Of course, not all countries that receive substantialaid grow rapidly. Some, like Namibia, have not. Further, there are examples of countries that havehad post-war GDP growth rates around or above 2 percent on a sustained basis without substantialODA (Vietnam for instance, and, more recently, El Salvador to a lesser extent).

On average in our sample, aid surges after conflict, but then falls to about 10 percent of GNI seven tonine years after conflict (Figure 4.15).For some countries, aid inflows continue to average 1520 percentof GNI for sustained periods (this high aid group includes Ethiopia, Mozambique, Namibia, theDemocratic Republic of the Congo, Nicaragua, Rwanda and Uganda). A sub-group that includesRwanda, Mozambique and Ethiopia dominates the sharp surge in growth of strong recovery countries.


5.0

10.0

15.0

20.0

25.0


Asapercentage

ofGNI



Figure 4.15 ODA flows: SGR versus WGR countries



26/38

Ugandas economy grew more slowly, but the aid inflows lagged as well. Mozambique and Rwandareceived the largest foreign aid spikes near the end of conflict: the upswing in aid started earlier inMozambique, averaging over 40 percent of GNI by 1990, two years before the war ended. Aid toMozambique surged again during the 2002 floods 10 years after the end of conflict. For Ethiopia andUganda, aid inflows rose after conflict ended (Figure 4.16).

Significantly, it appears that the pattern of growth recovery broadly mimics the flow of aid, with growthrecovering first in the countries that received the biggest doses of aid (Mozambique and Rwanda).Growth in Ethiopia and Rwanda started later and was uneven, but a late second surge of aid helped toboost Ethiopias growth rates after 2000. At the country level and for the group as a whole, greater aidinflows are broadly correlated with faster growth.78

What are the variables at play? First, grant aid helps to cover the budget deficit, sometimes to a verylarge extent. In some cases, countries have used aid to pull themselves out of slumps (Figures 4.17


Figure 4.16 GDP per capita growth and ODA as a share of GNI in selected countries

Ethiopia

0

2

4

6

8

10

12

14

16

18

20

E- 4 E- 3 E -2 E -1 En d E +1 E+ 2 E+ 3 E + 4 E + 5 E +6 E+ 7 E+ 8 E +9 E +10-10%

-8%

-6%

-4%

-2%

0%

2%

4%

6%

Asapercentag

eofGNI

GDPpercapita

growth

ODA (left axis)

Growth(right axis)

Mozambique

0

10

20

30

40

50

60

70

80

E -4 E -3 E-2 E -1 End E+1 E+2 E+3 E+4 E+5 E+6 E+7 E+8 E+9 E+10-3%

-1%

1%

3%

5%

7%

9%ODA (left axis)

Growth (right axis)

AsapercentageofGNI

GDPpercapita

growth


Rwanda

0

10

20

30

40

50

60

70

80

E-4 E-3 E-2 E-1 E nd E+1 E+2 E+3 E+4 E+5 E+6 E +7 E +8 E+9 E+10- 11%

-6%

-1%

4%

9%

14%

AsapercentageofGNI

GDPpercapitagrowth

ODA (left axis)

Growth(right axis)

0

5

10

15

20

25

E- 4 E-3 E -2 E -1 End E+1 E +2 E +3 E+4 E +5 E +6 E+7 E+8 E +9 E +100%

1%

2%

3%

4%

5%

6%

AsapercentageofGNI

Uganda

GDPpercapitagrowth

ODA (left axis)

Growth (right axis)


27/38


28/38

ground. Further, as indicated above, high inflation is not incompatible with high growth and theSGR countries ran larger public sector deficits for at least a few years after war.80

Aid may also have had an impact on economic performancethrough more indirect channels. Using aid to build social consen-

sus enables and fosters local policy initiatives, beyond merelyrepairing the damage wrought by war. Inclusive growth policiesmay have been made easier by aid.As shown below, SGR countriesdo receive on average more aid immediately after conflict ends andtheir policies have tended to be more inclusive. Such a deduction

would be compatible with the interpretation of the success of theMarshall Plan provided above. The political benefits of more andearlier aid may well be as important as its direct economic benefits.

Another important consideration for the use of aid is the rela-tionship between aid inflows and currency appreciation throughthe phenomenon known as the Dutch disease.81 However, therelationship is far from obvious. A study found that in 30 out ofthe 36 post-conflict countries considered, post-conflict aid inflows were not associated with realexchange rate (RER) appreciation, and sometimes the RER depreciated instead.82The study exam-ines seven cases in more detail, finding that even large surges in aid into Ethiopia, Uganda andRwanda had almost no effect on the competitiveness: in fact the inflation-adjusted exchange ratedepreciated somewhat. The conclusions are clear:

The assessment of macroeconomic outcomes of aid in the seven country experiencessuggests that post-conflict countries see aid as an important source for both financ-ing post-conflict spending needs and strengthening public-sector financial positionsby saving part of aid inflows. The latter takes the form of reducing public debtand/or hoarding international reserves, which are ways to prevent larger RER appre-ciation. Therefore it does not come as a surprise that the evidence on the simpleassociation between aid and RER appreciation is very mixed in this sub-sample.Most post-conflict aid-recipient countries appear to have exercised the option of notfully absorbing or spending aid. All this suggests the absence of large-scale Dutchdisease during the post-conflict cycle.83

There were, however, examples of post-conflict RER appreciation, but they seem largely disconnectedfrom aid inflows. In El Salvador and Nicaragua and to a lesser extent in Burundi, the exchange rate didappreciate in the post-conflict period, and in all three cases post-conflict growth was disappointing. ButBurundi experienced its RER appreciation seven to eight years after conflict,when aid inflows had begunto ebb. In the case of Nicaragua, the exchange rate continued to appreciate in spite of a rapid and sharpcutback in aid (from a high of over 70 percent of GNI a year after conflict to less than 25 percent three

years after conflict). El Salvador did not experience a substantial post-conflict aid surge, with post-con-flict aid peaking at seven percent of GNI one year after the end of conflict and decreasing steadily to lessthan 1.5 percent of GNI nine years after conflict (see Data Appendices).84 Further, despite receiving larg-er inflows of aid than WGR countries on average, SGR countries did not face RER appreciationonthe contraryso that competitiveness and export growth did not suffer (Figures 4.11 and 4.15).


Using aid to build socialconsensus enables and fosters

local policy initiatives, beyond

merely repairing the damage

wrought by war. Inclusive

growth policies may also be

facilitated by aid.


29/38

The way aid was used mattered.Uganda, for instance, avoided aloss of competitiveness due to aidinflows the same way Italy did in1949: by not absorbing all of the

(large amounts of ) aid it received.Uganda refrained from using itsaid inflows to import more than itexported. Ugandas concern tokeep its exchange rate fromappreciating may have been dueto the fact that 75 percent of itspopulation lives in rural areas anddepends mainly on agriculture, itsprimary export sector, for itslivelihood.85

It is important to recognize that while larger aid inflows mayindeed pose both managementand macroeconomic challengesto receiving countries, they alsoprovide greater flexibility in find-ing uses that suit local needs.

Debt relief

Aid in the form of debt relief doesnot provide a fresh inflow ofexternal finance.86 It does, howev-er, provide some fiscal space forthe post-conflict state authoritiesby facilitating the lowering ofinterest rates, allowing for someincremental borrowing, and mak-ing it easier for subsequent aidinflows to be spent on recoveryrather than used on repayments.

War is typically associated with asubstantial increase in the stock ofdebt87 and possible debt overhang,88

with well-known pernicious con-sequences for macroeconomicprospects.The conceptual and tech-nical questions raised by debt relief,either in the form of rescheduling or


BOX 4.5 LIBERIA CLEARS ITS ARREARS WITH THE IFIS AND REACHES

HIPC DECISION POINT

On 18 March 2008, Liberia became the thirty-third country to reach its decision point and

to qualify for debt relief under the enhanced HIPC initiative. The decision followed Liberias

clearances of its long-standing arrears to the IMF and the World Banks IDA. It also came as

recognition and in support of the Governments steadfast pursuit of reforms over the last

two years and will provide the boost needed to address the remaining, monumental

challenges faced by this post-conflict country, according to Ms Obiageli Ezekwesili, Vice

President of the Africa Region of the World Bank. Under the terms of the agreement,

Liberia will receive interim debt relief from some creditors, but in order to qualify for irrev-

ocable debt relief at the completion point, Liberia is expected to maintain macroeconom-

ic stability ..., prepare and implement a Poverty Reduction Strategy for at least one year;

and implement pivotal reforms in the areas of governance, PFM [public financial manage-

ment], debt management, as well as other structural and social measure. Upon reaching

the HIPC completion point, Liberia will become eligible for World Bank assistance under its

MDRI, from the African Development Bank and (beyond HIPC assistance) from the IMF.

The specifics of the Debt Relief Operation were as follows:

I Liberias public and publicly guaranteed external debt was estimated at US$4.7 bil-

lion in nominal terms as of 30 June 30 2007.

I Debt relief under the enhanced HIPC initiative is estimated at US

pcer_chapt4

Documents